- 1 -

Surface properties and wear performances

of siloxane-hydrogel contact lenses

Michela Bettuelli, Silvia Trabattoni, Matteo Fagnola, Silvia Tavazzi*, Laura Introzzi, Stefano

Farris

M. Bettuelli, Dr S. Trabattoni, M. Fagnola, Dr S. Tavazzi*

University of Milano Bicocca, Department of Materials Science, Via Cozzi 53, I-20125

Milano (Italy)

Dr L. Introzzi, Dr S. Farris

University of Milan, Department of Food, Environmental and Nutritional Sciences

(DeFENS)–Packaging Lab, Via Celoria 2, I-20133 Milano (Italy)

E-mail: silvia.tavazzi@unimib.it

Fax: +39 02 64485400

Running Heads: surface properties of siloxane-hydrogel contact lenses and OSDI

- 2 -

ABSTRACT

The low surface roughness of disposable contact lenses made of a new siloxane-hydrogel

loaded with hyaluronic acid is reported, as studied by atomic force microscopy (AFM).

Before the wear, the surface is characterized by out-of–plane and sharp structures, with

maximum height of about 10 nm. After a wear of eight hours, evidence of two typical

morphologies is provided and discussed. One morphology (sharp-type) has a similar aspect as

the unworn lenses with a slight increase of both the height and the number of the sharp peaks.

The other morphology (smooth-type) is characterized by troughs and bumpy structures.

Wettability and clinical performances are also discussed, the latter deduced by the Ocular-

Surface-Disease Index (OSDI). The main finding arising from this work is the indication of

correlation between the change of the OSDI before and after wear and the lens surface

characteristics obtained by AFM.

Keywords:

- New siloxane-hydrogel material;

- Atomic Force Microscopy (AFM);

- Disposable contact lenses;

- Wettability;

- Wear of contact lenses and Ocular Surface Disease Index (OSDI).

- 3 -

Introduction

In the last years, the advent of siloxane-hydrogel (Si-Hy) materials brought a revolutionary

advance in the contact lens field.[1] The main advantage of these materials is the ability of

siloxane moieties to promote the transport of oxygen and carbon dioxide across the lens. As a

consequence, the use of Si-Hy has shifted the topics of research from hypoxic complication,

removed by high gas permeability, to hydration and surface properties, due to the typical

relatively poor wettability of Si-Hy. To enhance wettability, which is typically measured by

the contact-angle technique, different approaches have been pursued by researchers and by

manufacturers. One strategy is based on surface treatments, plasma coatings, or plasma

oxidation;[2-4] another way is the inclusion of an internal wetting agent.[5,6] Another important

aspect is the hydration of the lenses, which typically ranges from 20% to 50% for lenses made

of Si-Hy materials.[7] A promising approach to maintain a high level of hydration is the

inclusion of hyaluronic acid (HA), which is a natural anionic polyelectrolyte present in many

tissues of the human body; it has wide applications in medicine and biotechnology and it has

also been proposed as a remedy for the dry-eye syndrome.[8-16] Besides measurements of

hydration and surface contact angle, atomic force microscopy (AFM) has also been used in

recent years to characterize contact lens surface both in liquid and anhydrous conditions.[17-22]

Changes of surface roughness after wear have been studied for some hydrogels and Si-Hy

based contact lenses. Indeed, the surface of a contact lens is in direct contact with the tear film

and the conjunctiva, thereby investigating the surface properties turns to be a crucial point

due to the involved clinical implications. Different tests can be used to evaluate the clinical

performances, such as staining of the ocular surface, Tear film Break-Up-Time (TBUT),

Schirmer test, Ocular-Surface-Disease Index (OSDI), the latter being used to have

information on the ocular symptomatology.[23] For example, during the wear of contact lenses,

the OSDI has been reported to be correlated to dry-eye symptoms and to mucin fragmentation

on the surface of the lens.[24-26]

- 4 -

Based on the aspects discussed above, this paper is aimed at presenting (i) some properties of

both new and worn contact lenses made of a new Si-Hy polymer loaded with HA, including

the surface properties deduced by AFM and contact-angle measurements, (ii) relevant

differences observed in the surface morphology of worn lenses, and (iii) the correlations that

have been found among the surface physical properties and the clinical performances of the

lenses, the latter described by the ocular-surface-disease index (OSDI).

Materials and Methods

The HA-containing[27] contact lenses used in this work (St. Shine Optical Co., LTD, Taiwan)

have been kindly supplied by Safilens (Italy). Some characteristics of the lenses (power

ranging from -1.00D to -4.00 D) are reported in Table I. A limited number of contact lenses of

the same material, but without HA, has been supplied by Safilens (Italy), only for chemical

and physical analyses. These lenses were not used for wear. They were loaded with HA[27,28]

in our laboratory in a water solution of HA (2.0 mg/ml) and the content of HA in each lens

has been evaluated to be of few tens of µg. Some characteristics of the used HA, such as the

average molecular weight and the polydispersity index are reported in Table S-I of the

Supporting Information.

Surface morphology has been studied by using the atomic force microscope (AFM)

Nanoscope V MultiMode (Veeco). Measurements have been performed in air in tapping mode

using RTESP silicon tips (Veeco, spring constant 20-80 N/m, resonance frequency 287-346

kHz). The lenses have been extracted from the blister with tweezers, rinsed three times in

deionized water to remove residual packaging solution, and then mounted on a semispherical

glass holder mimicking the lens curvature radius. Images of the front surface of the lenses

have been recorded with the resolution of 512512 pixels and corrected by zero or higher

order polynomial background filters depending on the scan size. Root-mean-square roughness

- 5 -

(Rrms), skewness (Rsk) and kurtosis (Rku) have been measured for all the analyzed lenses: 10

lenses denoted as CLrins, taken from the blister and rinsed with deionized water; 5 lenses

denoted as CLss, preserved 12 hours in saline solution and rinsed with deionized water; 22

lenses denoted as CLworn, which were worn for 8 hours, preserved for 12 hours in saline

solution, and rinsed in deionized water (for the worn lenses, 4 images have been eliminated

from the AFM analysis due to AFM experimental artifacts); 5 lenses purchased from

Johnson&Johnson (Galyfilcon A, Acuvue Advance, -3.00D) denoted as CLGaly. Rsk is a

measure of the asymmetry of the height value distribution with respect to the mean value,

showing positive values when the distribution exhibits a tail on its right side and negative

values when the tail is on the left side. Therefore, the Rsk value can be related to the

predominance of peaks (Rsk>0) or to the abundance of troughs (Rsk<0). Rku is related to the

shape of the probability distribution, showing values higher than 3 for sharp distributions and

values lower than 3 for bumpy shaped distributions. In this case, Rku value can be related to

the abundance of sharp structures (Rku>3) instead of bumpy structures (Rku<3). For the AFM

analyses, it has not been possible to acquire proper images on lenses taken from the blister

without rinsing them in water, because of the presence of salts, causing strong interactions

with the AFM tip.

Contact angle analyses have been performed using an optical contact-angle apparatus (OCA

15 Plus – Data Physics Instruments GmbH, Filderstadt, Germany) equipped with a video

measuring system with a high resolution CCD camera and a high performance digitizing

adapter. The software SCA 20 (Data Physics Instruments GmbH, Filderstadt, Germany) has

been used for data acquisition. Contact angles determinations have been carried out on four

different batches: lenses directly taken from the blister (CLbl), CLrins, CLss, and CLworn. The

procedure behind each sample preparation before analysis has been carefully standardized. In

particular, each sample has been first blotted to remove excess liquid, mounted on the same

custom-made glass holder as seen for the AFM analysis, and then blotted again by means of a

- 6 -

microfiber cloth, in order to further remove the residual liquid throughout the lens top

surface.[29] The worn lenses were stored after wear in saline solution for 12 hours before the

contact-angle measurements. Measurements have been run according to the sessile drop

method, i.e. gently dropping a droplet of (3.0 ± 0.4) µl of Milli-Q water (18.3 M cm) onto

the surface of the contact lens, at (20 ± 1)°C and (35 ± 5)% relative humidity. All droplets

have been released from a height of 1 cm above the surface to ensure consistency between

each measurement. The contact angle value has been taken 10 s after the droplet placement, to

reach the equilibrium at the solid/liquid interface. To this purpose, a semi-automatic image

analysis procedure has been adopted. For each lens type, the final contact angle value and its

standard deviation have been obtained from at least five measurements taken from different

replicates. This also allowed to monitor the repeatability of the results. The surface tension of

the blister packaging solution has also been assessed using the same apparatus as above,

coupled with the SCA 22 software, according to the pendant drop method.

Our group of wearers was made of 11 subjects aged 23±3 years (5 females and 6 males).

Some subjects repeated the wear twice at the distance of weeks. 7 subjects are habitual

wearers of contact lenses and 4 subjects are occasional wearers. They have been instructed to

wear contact lenses for 8 hours, after 3 days without any contact lens wear. Before the wear,

the ocular condition was evaluated and graded by Efron Grading Scales; the horizontal

corneal diameter, the eyelid aperture, and the blink frequency have been measured; the eye

topography has been collected, the tear film stability was assessed using tear film break time

(TBUT) and non-invasive TBUT; the lower Tear Meniscus Height (TMH) was measured

using a slit-lamp biomicroscope without fluorescein to evaluate the tear volume. Before the

wear, the wearers have been asked to provide information about their eye symptoms and

comfort during the day before the wear, filling up a questionnaire.[23] The same questionnaire

was used after wear to provide information on the eye symptoms during the 8 hours of wear.

From each questionnaire, the OSDI score was deduced. As reported in the literature,[23] the

- 7 -

OSDI has been validated to be a reliable instrument for measuring the ocular surface disease

when the questions are referred to a period of one week. We have used this index even if, in

our case, the questions were referred to a period of one day, either the day before the wear or

the day of wear of the contact lenses. The single score ranges from 0 to 4 (never, rarely,

sometimes, often, always) and is expressed for eight possibly-felt ocular sensations and for

four possibly-experienced problems in various activities. In some cases, the answers to some

questions were not given. This possibility is taken into consideration in the calculation of the

score.[23] The sum of all scores is defined as the OSDI and it ranges from 0 (maximum

comfort) to 100 (maximum discomfort). All subjects provided informed consent.

Statistical significance of differences among measured quantities for the different types of

lenses has been evaluated by Student’s t statistic (t-tests). Significant difference is considered

to occur if the probability value (p) is lower than 0.05. When studying the correlation between

measured quantities, the linear correlation coefficient R between the two groups has been

calculated. Significant correlation is considered to occur if the correlation probability (pc) is

lower than 0.05.

Results

Our first analysis has concerned the surface morphological characteristics observed by AFM.

As shown in Figure 1, the surface morphology of CLrins lenses taken from the blister and

rinsed in deionized water appears regular and relatively flat. Figure 2 shows the AFM images

taken on CLss lenses. Before wear, the main morphological characteristics are similar as for

the CLrins lenses. Changes were observed after 8-hour wear, as shown in Figures 3 and 4. In

these cases, we have observed two typical morphologies, with different characteristics. One

type of samples (Figure 3) shows structures which are out-of-plane and sharp, with maximum

height equal to about 30 nm, to be compared with 10 nm for the unworn CLrins lenses (Figure

- 8 -

1a). These observations suggested us to denote this morphology as sharp-type. Considering

the total number of examined worn lenses, this morphology occurred in the 60% of the cases.

Figure 4 represents a totally different morphology, characterized by troughs and bumpy

structures. It was indicated as the smooth-type morphology and it was found in the other 40%

of the wearers. Overall, the AFM morphology of the sharp-type lenses is similar to that of the

unworn lenses, while the smooth-type lenses show a very different aspect. This is clearly

evident by comparing Figs. 1, 3, and 4. The differences between the two morphologies of the

worn lenses are also evident in the 3D images in Figure 5.

Quantitative results arising from the AFM analyses are summarized in Table II. First of all,

we underline that the coefficient of variation (defined as the ratio between the standard

deviation and the mean) is typically relatively high, ranging from about 0.2 up to 1. These

large coefficients derive from the spread of the data, combined in some cases to the relatively

low mean value. After averaging the results of several samples, the roughness of the CLrins

lenses on (5 µm x 5 µm) scan size has been found equal to (1.5 ± 1.3) nm. This value has

been compared with that recorded for another Si-Hy material without surface treatments,

namely Galyfilcon A, showing similar water content and oxygen transmission level as the

material under investigation. For Galyfilcon A, the roughness deduced from (5 µm 5 µm)

images has been measured to be (1.6 ± 0.3) nm, similarly as reported in the literature.[22] The

difference among the roughness of these two types of lenses is not statistically significant (p =

0.8252, Table S-II). Also the (10 µm x 10 µm) roughness of Galyfilcon A ((1.9 ± 0.4) nm) is

comparable with the measured value ((2.6 ± 2.2) nm) for our Si-Hy polymer (p = 0.3429,

Table S-II). Indeed, both materials have a very regular and homogenous surface. However,

Galyfilcon A presents a porous morphology, as shown in Figure S-1 of the Supporting

Information. For the CLss lenses, the measured AFM roughness is similar as for the CLrins

lenses (p = 0.8032, Table S-III). For the worn lenses, Rrms is larger for the sharp-type worn

lenses than for the smooth-type ones, with p values indicating statistical differences (p =

- 9 -

0.0170, Table S-IV), with acceptable coefficients of variation, notwithstanding the reduced

number of measured samples. The sharp-type lenses also show a larger Rrms than the unworn

CLrins lenses, with statistical significance (p = 0.0275, Table S-V).

As far as the values of Rsk and Rku are concerned, the comparison between CLrins and CLGaly

confirms their different morphology, notwithstanding the similar Rrms. Indeed, as reported in

Table S-II, the Rsk values for these two lenses are statistically different (p = 0.0145) and Rku

shows a tendency to be different (p = 0.1010, Table S-II), within the limit of the very large

coefficients of variation. The skewness Rsk for the CLrins lenses is positive (2.8 ± 2.4) and

indicates that the structures which dominate the surface are out-of–plane. The kurtosis (Rku

>3) suggests the presence of sharp structures. For Galyfilcon A, Rsk is close to zero, indicating

that the height distribution is symmetric with respect to its mean value. The CLss lenses show

sharp and out-of–plane structures (Rku > 3, Rsk > 0), with no significant differences with

respect to the CLrins, lenses (p = 0.3804 and p = 8162 for Rsk and Rku, Table S-III). For the

worn lenses, Rsk and Rku are larger for the sharp-type worn lenses than for the smooth-type

ones (p = 0.0001 and p = 0.0082, respectively, Table S-IV), but the coefficients of variation

are very large. The sharp-type lenses also show a larger Rsk compared with the unworn CLrins

lenses (with no clear statistical significance, but indicative since p = 0.0769, Table S-V),

while Rku is comparable, as confirmed by the very high p value (0.9493) in Table S-V of the

Supporting Information. This indicates that the sharp-type lenses are characterized by sharp

peaks, similarly as CLrins, but the wear has determined an increase of the number of these thin

structures. On the contrary, the Rsk and Rku values of the smooth-type lenses can be

reasonably considered different with respect to the CLrins, with no clear statistical

significance, but indicative (Table S-V). This indicates that the modifications of the surface

have produced larger area with comparable heights, instead of sharp and relatively high peaks.

We have also evaluated the lens wettability by contact-angle measurements before and after

wear (as can be seen by the images displayed in Figure 6). In this case, it has also been

- 10 -

possible to analyze lenses directly taken from the blister without any rinsing (CLbl), in

contrast to AFM. The mean values and the standard deviations for the four types of lenses are

reported in Table III. The repeatability of the results is supported by the coefficient of

variation, which is found to be within few % (8% for CLbl, 7% for CLss, 4% for CLrins and

CLworn).

The measured contact-angle for the CLbl lenses is statistically very different with respect to all

the other types of lenses taken into consideration, as observed in Table III, as confirmed by

the p values in Table S-VI of the Supporting Information, and as it can also be seen by the

images displayed in Figure 6. On the contrary, when comparing the other lenses, no clear

evidences of statistically significant differences are found. Only the p values (Table S-VI) for

CLworn/CLrins and CLworn/CLss are indicative of possible differences, which are however

relatively small.

The physical characterization has been accompanied by a clinical study in terms of

symptomatology declared by the wearers through the OSDI. The results are reported in Table

IV, corresponding to the wearer symptomatology both before and after the wear. Considering

all wearers, the mean of the OSDI after wear was equal to 7.8±5.6 (not reported in Table IV),

to be compared with 4.6±3.4 before wear (p = 0.0142). In general, the wearers declared a

good eye condition after 8-hour wear with some slight differences among them. To

investigate in depth a possible relationship between surface lens properties and clinical

performances, the change in the OSDI before and after wear ((OSDI), Table IV) and the

parameters deduced by the AFM analysis after the wear are compared in Figure 7 for each

wearer. Black diamonds and red squares indicate the data corresponding to the different

observed morphologies. The numbers in the labels indicate each single wearer. In some cases,

AFM images were not taken into consideration for the analysis because affected by

experimental artifacts (images of poor quality), so that few numbers are lacking; other

numbers are repeated twice (one for each eye), other numbers are present only once. First of

- 11 -

all, we notice that the observed morphology for the right eye is the same as the morphology

corresponding to the left eye of the same wearer after the same period of wear. The Rrms, Rsk,

and Rku values for the right and left eyes of the same wearer are also in reasonable agreement,

their difference being usually relatively small compared to the difference with the results for

other wearers. The difference between the Rrms for the two eyes of the same wearer ranges

from 4.8% (wearer 7) to 26.6% (wearer 8) with respect to the wearer mean value. Only in one

case, a large difference has been found (79.3% for wearer 9 with sharp morphology). Similar

considerations can be done for Rsk and Rku, as can be reasonably understood from Fig. 7. We

also underline that, for two wearers (6 and 9), the AFM analysis has been performed after two

different periods of wear. The same morphology has been observed twice for the wearer 6,

whilst the wearer 9 showed the smooth morphology the first time and the sharp-type one after

the second period of wear. From Fig. 7, we notice that there is a non-negligible correlation

among the (OSDI) and the AFM results. The first panel indicates a correlation between

(OSDI) and Rrms (R=0.63, pc=0.0049). The worst comfort is correlated to the largest

roughness. A good correlation has also been found among (OSDI) and Rsk (R=0.74,

pc=0.0007).

Similarly as Fig. 7, Figure 8 shows the measured contact angle as a function of the

corresponding (OSDI). Evidence of correlation is found (R=0.78, pc=0.0009). As far as the

contact-angle data are concerned, it was not possible to split the results into the two smooth-

type and sharp-type groups.

Discussion

Taking into consideration the AFM observations, we underline that the unworn lenses have a

homogenous morphology without micro-pores or stripes and a relatively low value of

- 12 -

roughness, comparable to the roughness of other Si-Hy lenses without surface treatments (i.e.

Galyfilcon A). The low roughness of the unworn lenses and the non-ionicity of the material

lead us to expect a low content of protein lachrymal residues after wear. Indeed, after 8 hours

of wear, no large structures are detected on the lens surface. Moreover, the typical protein

absorption band at 280 nm was not detected in the UV-visible absorption spectrum of the

worn lenses (Supporting Information), thus indicating no protein residues within the

sensitivity limit of the UV-visible technique. A detailed investigation of possible lipid

deposits would be interesting. However, this aspect has not been investigated in this context.

Indeed, lipid deposits on the surface have been demonstrated to determine a decrease of the

contact angle[30], which has not been measured on our lenses after wear. We focused the

attention on the surface morphology and roughness. The increase in roughness for the sharp-

type morphology after 8 hours of wear could be unworthy for a not-daily lens, but previous

studies[31] have demonstrated that a significant changing in roughness often occurs after only

20 minutes of wear and it shows a slight increase during the remaining wearing period. At the

same time, for some wearers a new morphology appears (the smooth type). In this latter case,

the surface shows a new structure with voids, that we tentatively explain as caused by a loss

of materials. The observed modifications after wear for our smooth-type lenses are very

similar to the ones reported for other lenses (Acuvue) before and after testing them with an

abrasion device (reported by Goldberg, et al.[32]). In this latter case, AFM and scanning

electron microscopy analyses showed that the surfaces of both worn lenses and lenses tested

with the abrasion device were significantly altered with ridges, valleys, and formation of

elevations, depressions, and distinct grooves, due to damage of the surface. These authors

considered a period of wear of some weeks and they concluded that the surfaces of their

contact lenses were subjected to abrasive damage by the cornea during wear. This damage

was considered to be an initial mechanism event, followed by formation of interfacial debris

and deposits. In our case, for 40% of wearers, we found this effect after 8 hours of wear. The

- 13 -

effect has been observed on the front surface of the lens, thus being attributable to abrasion by

the eyelid. However, the origin of these surface damage requires further analysis. It could be

related to the lens fitting, but we believe that the wear conditions (physiological and/or

environmental) must play a pivotal role. Indeed, in our study, it may occur that the same

wearer presents either one or the other type of morphology in different wear occasions. For

example, this occurred for subject 9, who showed a type of morphology after 8-hour wear

and, at a distance of few weeks, the other type of morphology.

As far as the contact-angle measurements are concerned, we have measured a higher contact

angle after rinsing the lens, this increase after washing being also reported in the literature[29]

by Maldonado-Codina and Morgan, who observed an increase from 66° to 96° for Galyfilcon

A material. As pointed out by these authors, this can be due to the removal of surface active

agents upon washing. We believe that the increase of the contact angle after rinsing is due to

the removal from the surface of the lenses of active hydrophilic agents previously added in

order to enhance compatibility at the lens/eye interface by supporting a stable tear layer in the

eye. However, we are prone to exclude that this agent is HA. Indeed, in a previous study we

loaded hydrogel contact lenses with HA and we found that, after 5 days of immersion in 1 ml

of deionized water at 24 °C, the HA released by the lens was lower than 2.0 g/ml.[28] The

same effect arising from rinsing has been apparently achieved by soaking the contact lenses in

a saline solution. Also in this case, the increase of the contact angle is presumably due to the

disappearance of hydrophilic agents from the lens surface. It seems that the wear for 8 hours

has induced a further increase in the final contact angle with respect to CLrins and CLss lenses,

although with relatively poor statistical significance. Based on the data reported in the

literature, we stress that an increase of the contact angle is expected to be incompatible with

lipid deposits on the surface.[30] Our tentative explanation of the increase of the contact angle

is the modification of the surface morphology after wear. As also discussed in the

literature,[33-35] the morphology, the pattern, and the roughness of a surface are relevant

- 14 -

aspects in dictating the ultimate wettability properties (i.e. the contact angle values).

Moreover, loss of moisture at the lens/air interface should be taken into account.

The results of the physical characterization (AFM results and contact angles) have been

compared with the results of the clinical study. The OSDI values and their variations (OSDI)

are all relatively low, thus indicating a general good comfort. Within the small changes,

indications of correlation between the AFM parameters and (OSDI) are found, similarly as

also found between the contact angle and (OSDI). We also notice that the two morphologies

are distinguishable in Fig. 7. With the exception of the wearer 2, the points for the smooth

morphology (black diamonds) typically correspond to both lower AFM values and lower

(OSDI) values than the data for the sharp morphology. The difference between Rrms, Rsk, and

Rku for the two groups has already been discussed. On the contrary, the difference among the

corresponding (OSDI) is not confirmed by the p values (p = 0.2173), mainly due to the

anomaly of the results for the wearer 2 compared to the results for the other wearers of the

same group.

Conclusions

A new Si-Hy material for disposable hydrophilic contact lenses treated with HA has been

characterized. In general, the surface appears regular and flat. Before the wear, it is

characterized by out-of–plane and sharp structures, with maximum height equal to about 10

nm. After 12 hours in saline solution, the main morphological characteristics are preserved.

On the contrary, two typical morphologies are observed after 8 hours of wear. The former one

(sharp-type) has a similar aspect as the unworn lenses, even if the height of the sharp peaks

reaches about 30 nm and their number increases. The latter morphology (smooth-type) is

totally different. It is characterized by troughs and bumpy structures. In this case, we

- 15 -

attributed the changes to a loss of material, in agreement with the results of Goldberg et al.[32]

Typical low clinical OSDI values and (OSDI) values have been obtained. Indications of

different (OSDI) have been found corresponding to the two morphologies and indication of

correlation is found among (OSDI) and the parameters which describe the morphology of

the surface of the worn lenses, the worst clinical performances being correlated to the largest

Rrms and Rsk AFM parameters. This correlation can be explained as a consequence of the

surface damage of the lenses due to surface abrasion of the smooth-type lenses and due to the

changes of the surface parameters of the sharp-type lenses, even if the latter ones present a

similar morphology as the unworn lenses.

Also for the contact-angle data, experimental evidence of correlation with the clinical

performances has been found. As far as wettability is concerned, we have also found a strong

effect due to the rinsing with deionized water, which determines a remarkable increase of the

contact angle. In contrast to other explanations given in the literature, in our case this increase

has been attributed to the removal of active hydrophilic agents from the lens itself, rather than

to the blister solution. The slight further increase of the angle after wear is attributed to the

modification of the surface morphology after wear and possible loss of moisture at the lens/air

interface. Based on data reported in the literature, the increase of the contact angle is expected

to be incompatible with lipid deposits on the surface.

The expected impacts of these results are: (i) the exploitation of this new low-roughness

material for contact lenses, (ii) the further application of the well-known OSDI, whose

modification after wear is here demonstrated to be correlated to the surface morphology and

to the contact-angle measured on the worn lenses, (iii) the possible development of diagnostic

methods for the eye conditions, based on the correlation with the surface properties of the

worn lenses.

- 16 -

Acknowledgements

We acknowledge the approval of all the involved subjects, Alessandro Filippo, Alessandro

Borghesi, and Antonio Papagni for helpful discussions, and Peter Spearman for reading the

paper.

References

[1] Gonzales-Meijome JM, Lopez-Alemany A, Almeida JB, Parafita MA, Refojo MF.

Microscopic observations of superficial ultrastructure of unworn siloxane-hydrogel

contact lenses by cryo-scanning electron microscopy. J Biomed Mater Res Part B: Appl

Biomater. 2006;76B:419-423.

[2] Night&Day and AirOptix (Ciba Vision).

[3] PureVision (Bausch+Lomb).

[4] NanoglossTM PremiO (Menicon).

[5] Acuvue Oasys (Johnson&Johnson).

[6] Biofinity (CooperVision).

[7] Lin MC, Svitova TF. Contact lenses wettability in vitro: effects of surface-active

ingredients. Optom Vis Sci. 2010;87:440-446.

[8] Polack FM, McNiece MT. The treatment of dry eyes with Na hyaluronate-Healon.

Cornea 1982;1:133–136.

[9] Limberg MB, McCaa C, Kissling GE, Kaufman HE. Topical application of hyaluronic

acid and chondroitin sulfate in the treatment of dry eyes. Am J Ophthalmol.

1987;103:194-197.

[10] Hamano T, Horimoto K, Lee M, Komemushi S. Sodium hyaluronate eye drops enhance

tear film stability. Jpn J Ophthalmol. 1996;40:60–65.

- 17 -

[11] Yokoi N, Komuro A, Nishida K, Kinoshita S. Effectiveness of hyaluronan on corneal

epithelial barrier function in dry eye. Brit J Ophthalmol. 1997;81:533-536.

[12] Aragona P, Papa V, Micali A, Santocono M, Milazzo G. Long term treatment with

sodium hyaluronate-containing artificial tear reduces ocular surface damage in patients

with dry eye. Brit J Ophthalmol. 2002;86:181-184.

[13] Balazs EA. Viscoelastic properties of hyaluronan and its therapeutic use. In: Garg HG,

Hales CA, editors. Chemistry and biology of hyaluronan. Amsterdam: Elsevier.

2004:415.

[14] Falcone SJ, Palmeri D, Berg RA. Biomedical applications of hyaluronic acid.

Polysaccharides Drug Deliv Pharm Appl (ACS Symp Ser) 2006;155-174.

[15] Johnson ME, Murphy PJ, Boulton M. Effectiveness of sodium hyaluronate eyedrops in

the treatment of dry eye. Graefes Arch Clin Exp Ophthalmol. 2006;244:109-112.

[16] Kogan G, Solte´ s L, Stern R, Gemeiner P. Hyaluronic acid: a natural biopolymer with a

broad range of biomedical and industrial applications. Biotechnol Lett. 2007;29:17–25.

[17] Baguet J, Somer F, Duc TM. Imaging surfaces of hydrophilic contact lenses with atomic

force microscopy. Biomaterials. 1993;14:279-284.

[18] Teichroeb JH, Forrest JA, Ngai V, Martin JW, Jones L, Medley J. Imaging protein

deposits on contact lens materials. Optom Vis Sci. 2008;85:1151-1164.

[19] Giraldez MJ, Serra C, Lira M, Elisabete M, Real Oliveira CD, Yebra-Pimentel E. Soft

contact lens surface profile by atomic force microscopy. Optom Vis Sci. 2010;87:E475-

E481.

[20] Baguet J, Somer F, Claudon-Eyl V, Duc TM. Characterization of lachrymal component

accumulation on worn soft contact lens surface by atomic force microscopy.

Biomaterials. 1995;16:3-9.

- 18 -

[21] Gonzalez-Meijome JM, Lopez-Alemany A, Almeida JB, Parafita MA, Refojo MF.

Microscopic observation of unworn siloxane-hydrogel soft contact lenses by atomic

force microscopy. J Biomed Mater Res Part B: Appl Biomater. 2006;76B:412-418.

[22] Lira M, Santos L, Yebra-Pimentel E, Elisabete M, Real Oliveira CD. Comparative study

of silicone-hydrogel contact lenses surfaces before and after wear using atomic force

microscopy. J Biomed Mater Res Part B: Appl Biomater. 2008; 85B: 361-367.

[23] (a) Ocular Surface Disease Index (OSDI), Allergan Inc, Irvine, California (Italian

translation); (b) Schiffman RM, Christianson MD, Jacobsen G, Hirsch JD, Reis BL.

Reliability and validity of the Ocular Surface Disease Index. Arch Ophthalmol.

2000;118:615-621.

[24] Sengor T, Kurna SA, Ozbay N, Ertek S, Aki S, Altun A. Contact lens-related dry eye

and ocular surface changes with mapping technique in long-term soft silicone hydrogel

contact lens wearers. Eur J Ophthalm. 2012; 22:S17-23.

[25] Berry M, Purslow C, Murphy PJ, Pult H. Contact lens materials, mucin fragmentation

and relation to symptoms. Cornea 2012;31:770-776.

[26] Russo PA, Bouchard CS, Galasso JM. Extended-wear silicone hydrogel soft contact

lenses in the management of moderate to severe dry eye signs and symptoms secondary

to graft-versus-host disease. Eye Contact Lens 2007; 33:144-147.

[27] HA HyaCare (http://www.novozymes.com/en/investor/events-presentations/Document

s/3_CMD_BB_HA_HAOK_FINAL.pdf), Novozymes.

[28] Fagnola M, Pagani MP, Maffioletti S, Tavazzi S, Papagni A. Hyaluronic acid in

hydrophilic contact lenses: Spectroscopic investigation of the content and release in

solution. Cont Lens Anterior Eye. 2009;32:108-112.

[29] Maldonado-Codina C, Morgan PB. In vitro water wettability of silicone hydrogel

contact lenses determined using the sessile drop and captive bubble techniques. J

Biomed Mater Res Part A. 2007;83A:496–502.

- 19 -

[30] Lorentz H, Rogers R, Jones L. The impact of lipid on contact angle wettability. Optom

Vis Sci. 2007;84:946-953.

[31] Gonzales-Meijome JM, Lopez-Alemany A, Almeida JB, Parafita MA. Surface AFM

microscopy of unworn and worn samples of silicone hydrogel contact lenses. J Biomed

Mater Res Part B: Appl Biomater. 2009;88B:75-82.

[32] Goldberg EP, Bhatia S, Enns JB. Hydrogel contact lens-corneal interactions: a new

mechanism for deposit formation and corneal injury. CLAO J. 1997;23:243-248.

[33] Huh C, Mason SG. Effect on surface roughness on wettting. J Coll Interf Sci. 1977;

60:11-38.

[34] Synytska A, Ionov L, Minko S, Motornov M, Eichhorn KJ, Stamm M, Grundke K.

Tuning wettability by controlled roughness and surface modification using core-shell

particles. Polym Mater Sci Eng. 2004;90:624-625.

[35] Kubiak KJ, Wilson MCT, Mathia TG, Carval P. Wettability versus roughness of

engineering surfaces. Proc 12th Intern Conf Metr & Prop Eng Surf. Rzeszow, Poland.

July 8-10, 2009;265-275.

- 20 -

Figure captions

Fig. 1. AFM images with scan size 10 m x 10 m (a) and 5 m x 5 m (b) of the surface

morphology of lenses taken from the blister and rinsed in deionized water (CLrins).

Fig. 2. AFM images with scan size 10 m x 10 m (a) and 5 m x 5 m (b) of the surface

morphology of lenses preserved 12 hours in saline solution and rinsed with deionized

water (CLss).

Fig. 3. AFM images with scan size 10 m x 10 m (a) and 5 m x 5 m (b) of the sharp-type

surface morphology of lenses worn for 8 hours, preserved 12 hours in saline solution,

and rinsed in deionized water (CLworn).

Fig. 4. AFM images with scan size 10 m x 10 m (a) and 5 m x 5 m (b) of the smooth-

type surface morphology of lenses worn for 8 hours, preserved 12 hours in saline

solution, and rinsed in deionized water (CLworn).

Fig. 5. 3D AFM plots of lenses taken from the blister and rinsed in deionized water (CLrins)

(first panel) and worn for 8 hours, preserved 12 hours in saline solution, and rinsed in

deionized water (CLworn, second panel: smooth-type, third panel: sharp type).

Fig. 6. Exemplificative water droplet images captured 10 s after droplet deposition on the four

different types of lenses: (a) lens directly taken from the blister (CLbl), (b) lens taken

from the blister and rinsed in deionized water (CLrins), (c) lens preserved 12 hours in

saline solution and rinsed with deionized water (CLss), and (d) lens worn for 8 hours,

preserved 12 hours in saline solution, and rinsed in deionized water (CLworn).

- 21 -

Fig. 7. Roughness, skewness, and kurtosis deduced from AFM analysis of worn contact

lenses as a function of the change of the OSDI. Black diamonds correspond to the

smooth-type morphology. Red squares correspond to the sharp-type morphology. The

labels correspond to the subject identification numbers (repeated twice when data for

the right and left eyes are available, repeated more than twice when the wear has been

repeated at the distance of weeks). Continuous lines indicate the results of the linear

fitting of the data. The corresponding R and pc values are also given in each panel.

Fig 8. Measured contact angle of worn contact lenses as a function of the change of the

OSDI. A continuous line indicates the result of the linear fitting of the data.